“You can have overexcitation…but what about insufficient inhibition?” Martin said. “I think that is an equally important component.” Chang studied mice carrying the G93A mutant of the human superoxide dismutase 1 (SOD1) gene, which is associated with some forms of inherited ALS. The scientists are still looking for the specific spanner in the network. But whatever the cause of the disinhibition, Martin speculated that it could lead to a constant low level of activity—wearing out the motor neurons or poisoning them with their own excitotoxic emissions. The work suggests glycine signaling as a potential therapeutic target, said Mingchen Jiang of Northwestern University’s Feinberg School of Medicine in Chicago, Illinois, who was not involved with the study.

The balance of excitation and inhibition appears to be off in ALS. Excitatory spinal interneurons boost motor neuron firing in SOD1-G93A mice (see ARF related news story on Jiang et al., 2009). Motor neurons from the same model are particularly susceptible to excitotoxicity through AMPA glutamate receptors (Avossa et al., 2006). And in people, researchers have long known that those with ALS exhibit a reduction in glycine receptor levels (Hayashi et al., 1981; Whitehouse et al., 1983) as well as in glycine itself (Malessa et al., 1991; Niebroj-Dobosz and Janik, 1999).

Chang and Martin recently found that in SOD1-G93A mice, motor neuron innervation by inhibitory glycinergic interneurons, also called Renshaw cells, is reduced compared to control animals (Chang and Martin, 2009). In the current work, they extend their findings by performing neurophysiology on cultured neurons. First, they crossed the mSOD1 mice with animals expressing green fluorescent protein (GFP) under the Hb9, motor neuron-specific promoter. Single mutants with only the GFP marker served as controls. Chang cultured cells from embryonic spinal cords dissected from these mice and focused her attention on the green motor neurons.

The glycine receptor, upon glycine binding, allows chloride ions to flow into the cell and alter the membrane potential. Chang patch-clamped the cells and treated them with a dose of glycine, then measured the drop in current across the membrane. That drop in mSOD1 motor neurons was only two-thirds of the drop in control cells.

Chang also performed the experiments without adding glycine, analyzing the natural miniature inhibitory post-synaptic current (mIPSC) produced by glycine signals from the Renshaw interneurons in the culture dish. Again, the current changed less in the mSOD1 motor neurons. “The motor neurons are not getting the normal, spontaneous inhibition,” Martin said.

Because the spinal cultures contained a variety of cells, it was not possible to pin the blame solely on motor neurons with these experiments. The spontaneous inhibition defect could be a result of problems within the motor neurons themselves, in the interneurons that send glycine their way, or in any of the other cell types in the culture.

Martin and Chang designed an experiment to test mutant motor neurons in the presence of normal neighbors. Instead of culturing individual spinal cords, Chang isolated and pooled cells from wild-type and mSOD1 mouse cords together, creating a chimeric culture. Again, she performed patch-clamp studies with the fluorescently labeled motor neurons. This was a blinded study, because Chang was not able to genotype the neurons before the electrophysiology experiments. Instead, when the experiments were done, she sucked out the contents of individual cells and determined the SOD1 sequence.

Once the results were in, the scientists were surprised to find that glycine-induced current changes, as well as mIPSCs, were the same in mSOD1 and control motor neurons. That is, the healthy cells in the mixed culture somehow rescued the defect in the mutant neurons’ glycine response. The researchers are now investigating the possibility that the control interneurons innervate the mutant motor neurons. That hypothesis implies that at least part of the glycine defect is due to faults in the Renshaw interneurons.

But it appears that some fault lies with the mutant motor neurons, too. Chang counted glutamate receptor clusters on the surface of the motor neurons, and found that mutant cells exhibited approximately 20 percent fewer clusters, both in the soma and dendrites, than control neurons. And when Chang performed quantitative real-time reverse transcriptase polymerase chain reaction on the individual neurons, she found that mRNA levels for the glycine receptor were reduced by half in mSOD1 cells, suggesting the gene’s expression is downregulated. “These data, right now, are arguing that there could be both autonomous and non-autonomous mechanisms,” Martin said.

Martin and Chang are currently working to discover those. If glycine signaling is indeed part of the problem, then drugs that enhance the process could do some good. In fact, researchers at the Massachusetts General Hospital in Boston are planning an ALS trial of a drug that has that effect—the breast cancer drug tamoxifen.—Amber Dance.

Reference:
Chang Q, Martin LJ. Glycine receptor channels in spinal motoneurons are abnormal in a transgenic mouse model of amyotrophic lateral sclerosis. J Neurosci. 2011 Feb 23;31(8):2815-27. Abstract